Quantitative and Qualitative Trade-Off Analysis of Drowsy Driver Detection Methods: Single Electrode Wearable EEG Device, Multi-Electrode Wearable EEG Device, and Head-Mounted Gyroscope


Autoria(s): Chen, Emily; Durairaj, Dafydd; Hew, Bohr; Hoppel, Mark; Huang, Paula
Contribuinte(s)

Srinivasan, Aravind

Data(s)

08/06/2016

08/06/2016

01/05/2016

Resumo

Drowsy driving impairs motorists’ ability to operate vehicles safely, endangering both the drivers and other people on the road. The purpose of the project is to find the most effective wearable device to detect drowsiness. Existing research has demonstrated several options for drowsiness detection, such as electroencephalogram (EEG) brain wave measurement, eye tracking, head motions, and lane deviations. However, there are no detailed trade-off analyses for the cost, accuracy, detection time, and ergonomics of these methods. We chose to use two different EEG headsets: NeuroSky Mindwave Mobile (single-electrode) and Emotiv EPOC (14- electrode). We also tested a camera and gyroscope-accelerometer device. We can successfully determine drowsiness after five minutes of training using both single and multi-electrode EEGs. Devices were evaluated using the following criteria: time needed to achieve accurate reading, accuracy of prediction, rate of false positives vs. false negatives, and ergonomics and portability. This research will help improve detection devices, and reduce the number of future accidents due to drowsy driving.

Identificador

doi:10.13016/M25R3X

http://hdl.handle.net/1903/18079

Idioma(s)

en_US

Relação

Digital Repository at the University of Maryland

Gemstone Program, University of Maryland (College Park, Md)

Palavras-Chave #drowsy driver #Gemstone Team umdRoute #EEG headsets #detection devices
Tipo

Thesis